CUDA Application Design and Development

Chapter 9. Mixing CUDA and Rendering

With the focus on the use of CUDA to accelerate computational tasks, it is easy to forget that GPU technology is also a splendid platform for visualization. In combination with the Open Graphics Library (OpenGL), CUDA-enabled GPUs become visualization supercomputers. Even highly experienced OpenGL programmers will find this chapter both new and informative, as the examples use primitive restart—a feature recently added to the OpenGL 3.1 standard—to render high-performance, high-quality graphics even when the images require irregular meshes. Example kernels demonstrate data sharing between CUDA and OpenGL through buffer sharing. Profiling will show that primitive restart is 60 times faster than the optimized ...